Embodiments of the present invention relate to computer networks. In particular, embodiments of the present invention relate to an apparatus, method and computer program for connecting elements of a computer network together.
Connecting two computer network devices with data cabling correctly is difficult, as the network devices could be in any location and the wiring could be physically connected to more than one port on the network devices.
These problems are particularly prevalent on rack mounted storage enclosures connected via Serial Attached Small Computer System Interface (SAS) or Fibre Channel cabling, where the storage enclosures could be in any position on the enclosure rack, could have multiple identical external ports, and could have been originally physically wired incorrectly.
It is possible to wire up a set of enclosures incorrectly and use the connection available to determine in what way they are mis-configured. However, this is not always possible if the mis-wiring prevents exploratory gathering of connection data. Expanders cannot access the outside world through external Internet Protocol (IP) links, if the links are not wired or configured yet. Even if mis-wiring does allow for some correct gathering of system layout, it requires the system to be intelligent enough to make use of an incorrect set-up.
Software on a computer device could visually instruct and check the wiring of one or more server-room based machines. Usually, such automated checking is done through the wiring itself. This does not work if the wiring is either not connected or is mis-configured.